Capacity analysis of a diffusion-based short-range molecular nano-communication channel

Abstract

Simulation-based and information theoretic models for a diffusion-based short-range molecular communication channel between a nano-transmitter and a nano-receiver are constructed to analyze information rates between channel inputs and outputs when the inputs are independent and identically distributed (i.i.d.). The total number of molecules available for information transfer is assumed to be limited. It is also assumed that there is a maximum tolerable delay bound for the overall information transfer. Information rates are computed via simulation-based methods for different time slot lengths and transmitter-receiver distances. The rates obtained from simulations are then compared to those computed using information theoretic channel models which provide upper bounds for information rates. The results indicate that a 4-input-2-output discrete channel model provides a very good approximation to the nano-communication channel, particularly when the time slot lengths are large and the distance between the transmitter and the receiver is small. It is shown through an extensive set of simulations that the information theoretic channel capacity with i.i.d. inputs can be achieved when an encoder adjusts the relative frequency of binary zeros to be higher (between 50% and 70% for the scenarios considered) than binary ones, where a 'zero' corresponds to not releasing and a 'one' corresponds to releasing a molecule from the transmitter. (C) 2010 Elsevier B.V. All rights reserved.